Resected Lung Cancer Patients Who Would and Would Not Have Met Screening Criteria

Our plain-English summary of Resected Lung Cancer Patients Who Would and Would Not Have Met Screening Criteria, by F. Farjah, D.E. Wood, M.E. Zadworny, et al (The Annals of Thoracic Surgery, 2016).

The study

• Authors: F. Farjah, D.E. Wood, M.E. Zadworny, V.W. Rusch, N.P. Rizk
• Journal: The Annals of Thoracic Surgery
• Published: 2016;101(1):274-9
• Cited by: 3 articles, as counted by NIH iCite
• Read it: PubMed · Publisher · Free full text

Abstract

Background: Current eligibility criteria for lung cancer screening may underestimate the risk of malignancy for some individuals. We compared the predicted risk of lung cancer among patients who would have met screening criteria to those who would not have despite being at moderate-risk.

Methods: A retrospective cohort study of resected lung cancer patients was performed. The screen eligible group was based on criteria provided by the United States Preventive Services Task Force; age 55 to 80 and a 30 or greater pack-year smoking history. The screen ineligible group was based on criteria provided by the National Comprehensive Cancer Network for a moderate-risk individual not recommended screening; age greater than 50 years, greater than 20 pack-year smoking history, and no history of asbestos exposure or chronic obstructive pulmonary disease. A recently validated risk-prediction model was used to compare the risk of lung cancer across eligibility groups based on measured and imputed patient-level variables.

Results: Screen ineligible patients (n = 88) had a lower estimated probability of lung cancer than screen eligible patients (n = 419); 1.3% versus 3.1%, p value less than 0.001. However, 20% of screen ineligible patients had a predicted probability of lung cancer greater than or equal to the prevalence of lung cancer (3.7%) among National Lung Screening Trial participants; 17% of screen ineligible patients had a predicted probability of lung cancer greater than or equal to the American Association for Thoracic Surgery threshold (5%) defining high-risk individuals.

Conclusions: Current eligibility criteria for lung cancer screening underestimate the risk of lung cancer for some individuals who might benefit from lung cancer screening.

Copyright © 2016 The Society of Thoracic Surgeons. Published by Elsevier Inc. All rights reserved.

In plain English

This retrospective study examined patients who had lung cancer surgery at Memorial Sloan-Kettering Cancer Center, 1999 to 2008. It compared 419 who would have met US Preventive Services Task Force criteria (age 55 to 80, 30 or more pack-years) with 88 who would not have. Those 88 fit a National Comprehensive Cancer Network moderate-risk definition: age 50 or older, 20 or more pack-years, and no documented asbestos exposure or chronic obstructive pulmonary disease. Median model-predicted lung cancer risk was 1.3% in the ineligible group and 3.1% in the eligible group. In the ineligible group, 20% (95% confidence interval, 13% to 30%) had an estimated risk at or above 3.7%, the lung cancer prevalence among National Lung Screening Trial participants. In that group, 17% (95% confidence interval, 7% to 25%) reached the American Association for Thoracic Surgery's 5% high-risk threshold. The authors concluded that a substantial proportion of these moderate-risk patients would be considered high risk. They noted the study cannot show that screening such people would help, and that missing inputs were set identically for everyone.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Retrospective cohort study using a single-institution surgical database. Each patient's lung cancer risk was estimated with a previously published and validated risk-prediction model.
• Who was studied: Patients who had surgery for lung cancer at Memorial Sloan-Kettering Cancer Center between 1999 and 2008, with follow-up through 2012. 419 were screen eligible (age 55 to 80, 30 or more pack-years). 88 were screen ineligible (age 50 or older, 20 or more pack-years, no additional documented risk factor).
• Exposure: Groups were defined by age and smoking history (pack-years). Asbestos was not studied as an exposure. A documented history of asbestos exposure or chronic obstructive pulmonary disease, the other risk factors the database recorded, kept a patient out of the screen-ineligible group.
• Main results:
  • Median predicted probability of lung cancer was 1.3% (range 0.3% to 14%) in screen-ineligible patients and 3.1% (range 0.7% to 15%) in screen-eligible patients, p less than 0.001.
  • Among screen-ineligible patients, 20% (95% CI, 13% to 30%) had a predicted risk at or above 3.7%, the lung cancer prevalence among National Lung Screening Trial participants. The figure was 39% among screen-eligible patients (p = 0.001, Table 2).
  • Using the American Association for Thoracic Surgery 5% high-risk cutoff, 17% (95% CI, 7% to 25%) of screen-ineligible patients would have been considered high risk, compared with 22% of screen-eligible patients. This difference was not statistically significant (p = 0.319, Table 2).
  • Using the model's published thresholds for detecting 80% and 90% of lung cancers (1.6% and 0.9% in Table 2), 43% (95% CI, 33% to 54%) and 68% (95% CI, 57% to 78%) of screen-ineligible patients, respectively, would have been considered high risk.
  • Clinical and pathologic stage did not differ significantly between the groups. Carcinoid tumors were more common in the screen-ineligible group (5.7% vs 0.7%, p = 0.005). Squamous cell carcinomas were less common in that group (6.8% vs 19%, p = 0.004).
• The authors' conclusion: The authors concluded that many patients defined as moderate risk, and so not recommended for screening, would be considered high risk under several thresholds. In their view, current eligibility criteria may deny some high-risk individuals the benefits of early detection. They suggested that guideline groups consider recommending a risk-prediction model and defining a high-risk threshold.
• Limitations the authors state:
  • The study included only patients diagnosed incidentally on a diagnostic CT scan and treated for lung cancer. For that reason the authors cannot conclude that using a prediction model to select people for screening has benefits.
  • The study cannot identify an optimal risk threshold for screening.
  • Patients with similar predicted risk would not necessarily benefit equally from screening, for example because of other health conditions.
  • The authors cannot conclude that risk prediction would have led to earlier detection, because most patients already had early-stage disease without screening.
  • Model variables missing from the database were filled in with one common value for all patients: body mass index 27, some college education, no personal history of cancer, and no family history of lung cancer. The authors note that multiple imputation would have been a better approach.

Topics

The National Library of Medicine indexes this study under subjects our site covers: Lung cancer.

Medical terms in this study, explained in our Lexicon: Lung Cancer, Low-Dose CT Screening, Cohort Study, Asbestos Exposure, Thoracic Surgery.

Related studies

• Lung cancer risk in past asbestos workers a few decades after exposure cessation and prospects for screening — G. Mastrangelo, G. Marangi, et al. (2016)
• Prevalence and Preliminary Validation of Screening Criteria to Identify Carriers of Germline BAP1 Mutations — M.G. Zauderer, G. Jayakumaran, et al. (2019)
• Lung asbestos content in lungs resected for primary lung cancer — C.M. Wright, R.V. Bowman, et al. (2008)
• Current status of screening for malignant pleural mesothelioma — H.I. Pass, M. Carbone (2009)
• Screening for asbestos-induced diseases in Finland — K. Koskinen, J.P. Rinne, et al. (1996)
• Reactions of asbestos-exposed workers to notification and screening — B.E. Meyerowitz, C.D. Sullivan, et al. (1989)

Browse more research in The Annals of Thoracic Surgery, or return to medical research.